AI-based Network Anomaly Detection for CPS
exploiting 2D data representation
Mission of the project
ISEEYOO aims at designing and implementing an anomaly detection system for monitoring the information collected from a network of Cyber Physical Systems (CPSs) in the communication phase.
A CPS is a system characterized by a strong interaction among networking, sensing, and control functionalities. That is, the joint exploitation and coordination between computational and physical resources. The connectivity allowed by the use of the Internet, on one hand has extended the ability to share information and on the other has made these systems prone to vulnerabilities that did not exist before. The interaction between the Internet and CPSs requires greater efforts to ensure the security of connected systems.
In this project, we design a network anomaly detection system which exploits the distributed nature of CPSs by gathering the information coming from the set of deployed devices. To detect the presence of anomalies we adopt a 2D representation of traffic data thus characterizing the multi-input information collected from the sensors in a compact form.
The proposed approach is based on deep learning techniques and on an optimized 2D representation of the network data.
ISEEYOO will advance the state of the art by working towards three specific objectives:
Objective 1
Selection of the most effective two-dimensional representation of network traffic information
Objective 2
Definition of a deep learning model for the detection of anomalies based on the two-dimensional representation of traffic data
Objective 3
Design and implementation of a network anomaly detection system by exploiting the information gathered from the set of distributed nodes
To achieve these ambitious goals, ISEEYOO will exploit background knowledge from academia, technology experts and high-tech industry. Academic leaders in the disciplines of network security, image processing as well as machine learning and data processing techniques will participate in this research (UNIPD, UNIRM3, ICCS), together with a company active in the development of communication technologies or their integration into commercial products (TEI). To find out more about the ISEEYOO team take a look at the "Team" section.
Questions?
Contact federica.battisti@unipd.it to get more information on the project